JOURNAL OF CLINICAL MICROBIOLOGY, JUIY 1989, P. 1640-16450095-1137/89/071640-06$02.00/0Copyright (O 1989, American Society for Microbiology

Immunoblot Detection of Class-Specific Humoral Immune Responseto Outer Membrane Proteins Isolated from Salmonella typhi in

Humans with Typhoid FeverVIANNEY ORTIZ,1 ARMANDO ISIBASI,2* ETHEL GARCIA-ORTIGOZA,3 AND JESUS KUMATE4

Laboratorio de Amibiasis Experimental, Instituto Nacional de Higiene, Mexico City, D.F., Laboratorio deInmunoquiimica, Unidad de Investigacién Biômedicca, Instituto Mexicano del Segiuro Social, P.O. Box 73-032, MexicoCity, D.F. 03020,2 Laboratorio de Inmalnoqaiznica J, Departamnento de Ininunologia, Escuela Nacional de CienciasBiologicas, I.P.N., Mexico City, D.F.,' and Subsecretaria de la Secretaria (le Sallad, Mexico City, D.F.,4 Mexico

Received 19 November 1988/Accepted 22 March 1989

The studies reported here were undertaken to assess the ability of the outer membrane proteins (OMPs) ofSalmonella typhi to induce a humoral immune response in humans with typhoid fever. OMPs were isolated withthe nonionic detergent Triton X-100 and were found to be contaminated with approximately 4% lipopolysac-charide. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis patterns showed protein bands withmolecular size ranges from 17 to 70 kilodaltons; the major groups of proteins were those that correspond to theporins and OmpA of gram-negative bacteria. Rabbit antiserum to OMPs or to S. typhi recognized OMPs afterabsorption with lipopolysaccharide. Sera from patients with typhoid fever contained immunoglobulin Mantibodies which reacted with a protein of 28 kilodaltons and immunoglobulin G antibodies which reactedmainly with the porins, as determined by immunoblotting. These results indicate that the porins are the majorimmunogenic OMPs from S. typhi and that the immune response induced in the infection could be related tothe protective status.

Typhoid fever is an unsolved health problem in the world.There are approximately 12 million cases per year, and indeveloping countries the incidence is on the order of 540 newcases per 100,000 inhabitants (6). Eradication of the diseasein these countries depends mainly on improvement of sani-tary and nutritional conditions, but in the meantime, a goodvaccination program could help in the control of the disease.The main problem for the development of an effectivevaccine is that up to now little is known about the microbialfactors that determine pathogenicity and those that elicit a

protective immune response in humans with typhoid infec-tion.

Available vaccines containing either acetone or heat-killedSalmonella typhi are of limited value, not only because theyconfer short-lived protection but also because they produceunacceptable side effects, due mainly to the presence ofendotoxin, which prevent their use in children (8, 18).Germanier and Furer have developed an oral vaccine froma strain of S. typhi (Ty2la) which is deficient in the enzymeUDPgalactose epimerase (10). Although this vaccine lacksthe ill effects of the previous ones, it is expensive and hasshown a variable efficacy in the trials that have been done inEgypt (97%) and Chile (67%) (9, 32). At the moment thereare also several vaccines prepared from Vi antigen still in thephase of experimentation in humans (1).Up to now, the antigens of S. typhi that participate directly

in inducing protection are not known. In humans, thehumoral immune response against some superficial antigenshas been studied. In the sera of typhoid fever patients, thereare antibodies against 0, H, and Vi antigens (13, 28).Nevertheless, high titers of these antibodies have no rela-tionship to the state of protection or to the evolution of thedisease. That is why the search for specific antigens involvedin protection is still in progress.

* Corresponding author.

Due to the localization of the outer membrane proteins(OMPs) on the surface of gram-negative bacteria (25, 27),OMPs have been recently considered as important antigensin the induction of a specific, protective immune response.

OMPs from S. typhi have molecular sizes in the range from17 to 80 kilodaltons (kDa) (14). When they are isolated by themethod of Schnaitman (29), a contamination with lipopoly-saccharide (LPS) of 4% is obtained. We have demonstratedin a mouse model that these OMPs induce protection againsta challenge with S. typhi (15); the same protective effect wasobtained when mice were passively injected with rabbithyperimmune anti-OMP sera; when the rabbit antisera were

previously absorbed with OMPs this protective effect disap-peared, but when they were absorbed with LPS there was no

significant loss of protection. These results indicate that theOMPs are responsible for the protection observed in thisanimal model.

Considering the results mentioned above, it is important tofind out whether OMPs from S. typhi are also immunogenicin humans. For this reason we analyzed the immune re-

sponse to OMPs in patients with typhoid fever. In this work,we demonstrate that the humoral immune response of suchpatients is directed mainly to porins in the convalescentphase of the disease.

MATERIALS AND METHODS

Bacterial strains. The virulent strain S. typhi 9,12,Vi:d(hereafter referred to as M01) was isolated from a patientwith typhoid fever and has been maintained in culture in our

laboratory since 1979. S. typhi 0-901, Ty2, and Ty2la andSalmonella typhimurium 1,4,5,12 were kindly donated bythe Instituto Nacional de Higiene, Mexico City, Mexico.

Isolation of outer membrane proteins. OMPs were obtainedas described by Schnaitman (29). Briefly, S. typhi M01 was

grown in minimum salts medium (medium A) containing0.5% yeast extract and 1.25% glucose and incubated in a

1640

Vol. 27, No. 7

ANTIBODIES AGAINST PORINS IN TYPHOID FEVER 1641

shaker at 37°C for 8 h to mid-exponential growth to anoptical density of 0.36 at 660 nm (previously determined bya growth curve). Cells were harvested and suspended in 0.01M HEPES (N-2-hydroxyethylpiperazine-N'-2-ethanesulfo-nic acid) buffer. Bacteria were treated with an OmniMixer(Ivan Sorvall, Inc., Norwalk, Conn.) to remove flagella andthen were disrupted by sonication, using an Ultratip Lab-sonic system (Lab-Line, Melrose Park, Iii.), and centrifugedat 7,000 x g for 10 min to remove intact cells. The superna-tant fluid was centrifuged at 200,000 x g for 45 min at 4°C toobtain the cell envelope. The cytoplasmic membrane wasremoved with 0.01 M HEPES containing 2% Triton X-100.followed by centrifugation at 200,000 x g for 45 min at 4°C.to achieve solubilization of the OMPs, the Triton X-100-insoluble fraction (outer membrane and peptidoglycan) wassuspended in 0.05 M Tris containing 5 mM EDTA and 2%Triton X-100, pH 7.4, and was allowed to stand 10 min at37°C. Finally, the suspension was centrifuged as describedabove but at 37°C, and the OMPs were recovered in thesupernatant fraction. Protein concentration was determinedby the method ofLowry et al. (24). LPS contamination of theprotein samples was 4%, as determined indirectly by mea-suring the concentration of 2-keto-3-deoxyoctulosonic acid(20); as a standard, 2-keto-3-deoxyoctulosonic acid obtainedfrom Sigma Chemical Co., St. Louis, Mo., was employed.

Isolation of LPS. S. typhi 0-901 LPS was extracted by thephenol-water method (34).Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

(SDS-PAGE) and immunoblotting. SDS-PAGE was per-formed under reducing conditions by using the discontinu-ous buffer systems of Laemmli (23) with a vertical-slab gelelectrophoresis unit (Hoefer Scientific). The separating gelcontained 11.2% acrylamide, 2.5% bisacrylamide, and 0.19%SDS on 0.35 M Tris hydrochloride buffer (pH 8.8). Thestacking gel contained 5% acrylamide, 0.13% bisacrylamide,and 0.1% SDS in 0.12 M Tris hydrochloride buffer (pH 6.8).The running buffer was 0.025 M Tris hydrochloride-0.192glycine (pH 8.3) with 0.1% SDS. Electrophoresis was per-formed at 30 mA per plate.

Electrophoretic transfer ofOMP from polyacrylamide gelsto nitrocellulose paper (NCP) was accomplished in an elec-troblotting unit (Hoefer Scientific) by means of the transferbuffer described by Towbin et al. (31). The proteins wereelectrophoresed at 100 mA for 18 h. NCPs were immersedfor 1 h at 37°C in blocking buffer (0.15 M NaCl, 0.01 Mphosphate buffer [pH 7.2], 0.25% gelatin, 6 mM EDTA).After being washed in 0.15 M NaCl-0.01 M phosphate buffer(pH 7.2)-0.1% Tween 20 (PBS-T), the NCPs were incubatedfor 3 h at 25°C in a 1:100 dilution of the test serum samplesin blocking buffer. After being washed in PBS-T, the NCPswere incubated for 1.5 h at 25°C with optimal concentrationsof horseradish peroxidase-conjugated goat anti-rabbit immu-noglobulin G (IgG) or anti-human IgM or IgG (CopperBiomedical, Inc., West Chester, Pa.). After two washes inPBS-T and one in phosphate-buffered saline only, the NCPswere incubated in substrate solution (2 mM 4-chloro-2-naphthol and 0.08% HO, in phosphate-buffered saline) for10 min. After a wash in tap water, the NCPs were photo-graphed, air dried, and kept in the dark.Human sera. Sera from eight persons (ages between 25 and

40) with clinical and bacteriological diagnosis of typhoidfever, hospitalized at the Hospital de Enfermedades Infec-ciosas, Centro Médico la Raza, IMSS, Mexico City, Mexico,were analyzed for IgM and IgG responses to OMPs in boththe acute (1 day after hospital admission) and the convales-cent (30 days after release from the hospital) phases. Sera

A B C

66 Kd

45 Kd

-r -

oe

29 <d

14 Kd

FIG. 1. SDS-PAGE of S. tîphfi MOl OMPs. Lanes: A, solublefraction in HEPES-Triton X-100: B, insoluble fraction in Tris-EDTA-Triton X-100 (peptidoglycan); C, soluble fraction in Tris-EDTA-Triton X-100 (OMPs). Molecular size markers: bovine serumalbumin (66.2 kDa), ovalbumin (45 kDa), carbonic anhydrase (29kDa). and lysozyme (14.4 kDa).

from 10 persons without previous known contact with S.typhi were used as controls.Anti-OMP antisera. Anti-OMP sera were raised in New

Zealand White rabbits (2.5 to 3 kg) intradermally injectedwith different concentrations of OMPs; the first two injec-tions (1 mg/ml) were given in Freund complete adjuvant ondays 0 and 7. After day 15, the rabbits received three furtherinjections of 0.15, 0.5, and 1.0 mg/ml given every other daywithout Freund complete adjuvant. The rabbits were bled 10days after the last injection.

RESULTS

OMP preparation. S. typhi M01 OMPs isolated as de-scribed above were contaminated with 4% LPS. Figure 1shows the SDS-PAGE electrophoretic pattern. As seen,there are differences between fractions obtained during theprocess of purification. The fraction soluble in HEPES-Triton X-100 (lane A), which corresponds to the cytoplasmicmembrane proteins, shows more than 15 bands, most ofthem different from those observed in the other two fractions(lanes B and C). The insoluble fraction in Tris-EDTA-TritonX-100 (lane B), which corresponds to peptidoglycan, pre-sents mainly three proteins of molecular sizes around 36 to41-kDa. The fraction soluble in Tris-EDTA-Triton X-100(lane C) corresponds to the OMP preparations. It presentssome minor proteins with molecular sizes ranging from 17 to70 kDa and the characteristic group of major proteins of 36 to41 kDa, as well as the 28- and the 17-kDa bands.To confirm whether we really had obtained OMPs from S.

tyvphi M01, we compared PAGE patterns of similar prepara-tions obtained from other strains of S. typhi as well as fromS. tîphimuriun, which has been well characterized by other

VOL. 27. 1989

1642 ORTIZ ET AL.

A B C D E F,*~~~ ~ ~ ~ ~ ~ ~ ~ ~

Mo

4._FIG. 2. SDS-PAGE of OMPs from S. typhi M01, 0-901, Ty2,

and Ty2la and from S. typhimuriumn (lanes B, C, D, E, and F.respectively). Molecular size standards: bovine serum albumin (66.2kDa), ovalbumin (45 kDa), carbonic anhydrase (29 kDa), andlysozyme (14 kDa) (lane A, from top to bottom, respectively). Thearrowhead indicates the porins.

authors (19). In Fig. 2, we observe that the S. typhi M01pattern (lane B) is very similar to those of S. typhi 0-901(lane C), Ty2 (lane D), and Ty2la (lane E) and S. typhimu-rium (lane F). All of them have the major proteins of 36 to 41kDa that are called porins.Immunoblotting with rabbit anti-OMP sera. Sera from

rabbits immunized with S. typhi OMPs were analyzed byimmunoblot to determine the antigenicity of these proteins.The immunoblot analysis is shown (Fig. 3). Anti-OMP rabbitimmune sera reacted with greater intensity against the pro-tein bands located between 36 and 41 kDa, which correspondto the so-called porins and OmpA. They also reacted againstthe other OMPs (Fig. 3, lane la). This pattern of recognitionwas not modified, in comparison with the number of bandsseen when the antisera were previously absorbed, at equiv-alence, with LPS from S. typhi (Fig. 3, lane lb). Rabbitsimmunized with whole bacteria produced antibodies againstOMPs (Fig. 3, lane 2a); when the sera were absorbed withLPS (Fig. 3, lane 2b), no changes were observed. Neverthe-less, the intensity of the reaction was much less because theantisera were diluted two times during the absorption pro-cedure. The results demonstrate that the antibodies aredirected against OMPs and not against LPS.Immunoblotting with patient or normal human sera. To

determine the antigenicity of the OMPs from S. typhi inhumans, antibodies against these proteins were detected, byimmunoblotting, in sera of patients with typhoid fever, inboth the acute and the convalescence phases. The analysisof the primary and secondary responses showed the follow-ing results.

(i) IgM class response. Four of eight patients (patients 1, 4,5, and 8) had antibodies of the IgM class which recognized aprotein with a molecular size of 28 kDa during the acutephase; only two of these (patients 5 and 8) showed the sametype of response during the convalescence phase (Fig. 4).The 28-kDa protein was the only one recognized by thepatient antibodies. None of the 10 healthy controls showedany type of response (data not shown).

1

a b

66.2K-K

45O0 K-

2a b

14.4K-

FIG. 3. Immunoblotting of OMPs from S. typhi M01 detected byrabbit antibodies and peroxidase-conjugated goat anti-rabbit poly-valent immunoglobulin. Lanes: la, rabbit anti-S. typhi OMPs; lb,rabbit anti-S. typhi OMPs absorbed with LPS; 2a, rabbit anti-wholeS. typhi; 2b, rabbit anti-whole S. typhi absorbed with LPS. Absorp-tion with LPS was done at equivalence. Nonabsorbed sera wereused at a 1:100 dilution, absorbed sera were used at a 1:200 dilution,and the conjugate was used at a 1:1,000 dilution. Molecular weights(in thousands) are given at the left.

(ài) IgG class response. The secondary antibodies present intyphoid fever patients were directed mainly to the majorOMPs, that is, to the 17- and 28-kDa proteins and to theporins (Fig. 5). The response to the 17-kDa protein waspresent in only three of eight patients (patients 1, 2, and 5) inboth the acute and the convalescence phases. To the 28-kDaprotein, four of eight patients (patients 1, 2, 4, and 5)responded in the same way in the acute and the convales-cence phases. During this latter phase, one other patient(patient 3) also showed a positive reaction. Response to theporins was more evident since during the acute phase four ofeight patients (patients 1, 2, 4, and 5) gave an intensereaction that was also present during the convalescencephase. In this latter phase, three more patients (patients 3, 6,and 8) had antibodies which recognized the porins. Only 1 of10 control sera recognized the porins and the 28-kDa protein(Fig. 6, lane a).

DISCUSSION

In the present work, we have demonstrated that OMPsfrom S. typhi M01, a strain isolated from a typhoid feverpatient, presents an SDS-PAGE pattern similar to those ofother strains of S. typhi, such as Ty2 (parenteral vaccinestrain), Ty2la (oral vaccine strain), and 0-901 (nonflagel-lated strain). The proteins that are common in all strains arethe 17-, 28-, and 36- to 41-kDa proteins. The pattern for S.typhimurium, which is very similar to that reported in theliterature (19), allowed us to identify the major proteins,including the 36- to 41-kDa proteins which are called porins.By immunoblot analysis, we showed that the OMPs of S.

typhi M01 are immunogenic in rabbits when they are admin-istered either in isolated form or as a component of thebacteria. The IgG anti-OMP antibodies recognized the majorproteins present in the preparations, including the porins.

J. CLIN. MICROBIOL.

imrm-k--z.>c4:::- i"& lm -.c

lllllp.W.' ',

ANTIBODIES AGAINST PORINS IN TYPHOID FEVER 1643

2 3 4a c a c a c

5 e sa c a c a c a c

66.2K..-

1K.14C.5OK--

14.4K- |-*.-

FIG. 4. IgM responses of eight patients with typhoid fever (panels 1 to 8) against OMPs from S. typhi, detected by immunoblotting. Serawere diluted 1:50, and peroxidase-conjugated goat anti-human IgM was diluted 1:1.000. Lanes: a, acute phase; c, convalescence phase.Arrows indicate the proteins recognized by sera of patients. Molecular weights (in thousands) are shown.

This reactivity was not essentially modified when anti-LPSantibodies were previously absorbed, which shows the im-portance of proteins as immunogens.We have also demonstrated that S. typhi M01 OMPs are

immunogenic in typhoid fever patients. The immunoblotanalysis performed permitted us to identify which proteinsare recognized by IgM and IgG antibodies from patients in

the acute and the convalescence phases of the disease. We

1 2 3a c a c a c

66.2K -

45.0K4q

4

a c

observed that four of eight patients had IgM antibodiesagainst the 28-kDa protein during the acute phase. In theconvalescence phase, only two of these patients continuedto present this reactivity. We did not observe any reaction tothe porins. This is in disagreement with the results obtainedby Calderon et al. (4) using enzyme-linked immunosorbentassay. This difference could be related to the nature of theantigen used in both works. In our experiments the OMPs

5 6 7 8

a c a c a c a c

66.2 K -.4

w

45.0K--»

1 4.4K -t_àe14.4 K -

FIG. 5. IgG responses of eight patients with typhoid fever (panels 1 to 8) against OMPs from S. tîphi, detected by immunoblotting. Serawere diluted 1:50, and peroxidase-conjugated goat anti-human IgG was diluted 1:1.000. Lanes: a. acute phase; c, convalescent phase. Arrows

indicate the proteins recognized by sera of patients. Molecular weights (in thousands) are shown.

aa C

66.2 K -

45.0K-

14.4K-

VOL. 27, 1989

1644 ORTIZ ET AL.

G d e f

66.2-

45.0-_> M

14A4-

FIG. 6. Immunoblotting of OMPs from S. typhi detected in sera

from 10 healthy persons (lanes a to j). Sera were diluted 1:50, andperoxidase-conjugated goat anti-human IgG was diluted 1:1,000.Arrows indicate the proteins recognized by sera. Molecular weights(in thousands) are shown.

were denatured, whereas Calderon and co-workers usedporins in a native form; therefore, it is possible that IgManti-porin antibodies react mainly with the proteins in a

native form. The IgG response was different from theprimary one, since several proteins were recognized in boththe acute and the convalescence phases. It is important topoint out that the response to porins was the most relevantbecause an intense reaction was observed in four of eightpatients in the acute phase and in seven of eight patients inthe convalescence phase. These results indicate that porinsare the immunodominant antigens in these OMP prepara-

tions, which is in agreement with the results of Calderon etal. (4), who observed high antibody titers to porins intyphoid fever patients.The ability of OMPs to induce a protective immunity in

infection caused by diverse gram-negative bacteria, such as

Haemophilus influenza type b (12), Neisseria meningitidisgroup B (33), Pseudomonas aeruginosa (11), S. typhimurium(22), and Bordetella bronchiseptica (26), has been demon-strated previously. The importance of the production ofspecific anti-OMP antibodies for protection has been sug-gested also for infections caused by P. aeruginosa (7),Campylobacterjejuni (2), and Chlamydia trachomatis (5).For S. typhi 9,12,Vi:d, active immunization of mice with

its OMPs induced protection against a challenge with thebacteria (15). The same effect was observed with passiveadministration of a rabbit hyperimmune serum againstOMPs. This same serum induced partial protection against a

challenge with S. typhimurium. This cross-reactive protec-tion was attributed to the antibodies against the porins (15).

In typhoid fever, the presence of high titers of antibodiesagainst the somatic O antigen has no relationship to protec-tion. This could be because the antigen-antibody reactiontakes place in the most external portion of the somaticantigen, that is, in the oligosaccharide units of repetition,and when complement is activated the C5-C9 membraneattack complex (MAC) cannot be inserted in the membraneto promote bacterial lysis (16, 17). On the other hand,anti-porin antibodies recognize antigenic determinants ex-

h posed directly on the outer membrane surface, which allowsthe insertion of MAC and causes damage. This could be the

; natural way of controlling the disease during bacteremia.Nevertheless, it is important to remember that S. typhi is anintracellular germ; for that reason, the cellular effectormechanisms are also very important in the control of thedisease. Because production of antibodies against proteins isa T-cell-dependent immune response, porins must activate Tcells (CD4) which will then release lymphokines specific forB cells, (IL4, ILS, and IL6) (21), T cells (IL2) (30), andmacrophages (INF--y) (3). The interactions of all these cells,lymphokines, and antibodies must generate a protectiveimmune status. It is important that preliminary results haveshown that porins from S. typhi stimulate in vitro-prolifera-tive immune responses in murine immune T lymphocytes (J.

* Moreno et al., manuscript in preparation).In conclusion, we can say that porins probably play an

important role in protective immunity against typhoid fever,since they stimulate both the humoral and cellular effectormechanisms.

.,....

LITERATURE CITED1. Acharya, I. L., C. L. Lowe, R. Thapa, V. L. Gurubacharya,

M. B. Shresta, D. Bact, M. Cadoz, D. Schulz, J. Armand, D. A.Bryla, B. Trolifors, T. Cramton, R. Schneerson, and J. B.Robbins. 1987. Prevention of typhoid fever in Nepal with the Vicapsular polysaccharide of Salmonella typhi. A preliminaryreport. N. Engl. J. Med. 317:1101-1104.

2. Blaser, M. J., J. A. Hopkins, and M. L. Vasil. 1984. Campylo-bacterjejuni outer membrane proteins are antigenic for humans.Infect. Immun. 43:986-993.

3. Buchmeier, N. A., and R. D. Schreiber. 1985. Requirement ofendogenous interferon-gamma production for resolution of List-eria monocytogenes infection. Proc. Nati. Acad. Sci. USA82:7404-7408.

4. Calderon, I., S. R. Lobos, H. A. Rojas, C. Palomino, L. H.Rodriguez, and G. Mora. 1986. Antibodies to porin antigens ofSalmonella typhi induced during typhoid infection in humans.Infect. Immun. 52:209-212.

5. Caldwell, H. D., and L. P. Perry. 1982. Neutralization ofChlamydia trachomatis infectivity with antibodies to the majorouter membrane protein. Infect. Immun. 38:745-754.

6. Edelman, R., and M. M. Levine. 1986. Summary of a workshopon typhoid fever. Rev. Infect. Dis. 8:329-349.

7. Fernandez, P. B., C. Kim, K. R. Cundy, and N. N. Huang. 1981.Antibodies to cell envelope proteins of Pseudomonas aerugi-nosa in cystic fibrosis patients. Infect. Immun. 33:527-532.

8. Germanier, R. 1984. Typhoid fever, p. 135-165. In R. Ger-manier (ed.), Bacterial vaccines. Academic Press, Inc., NewYork.

9. Germanier, R. 1986. The live oral typhoid vaccine Ty2la: recentfield trial results, p. 10-12. In Sclavo International Conferenceon Bacterial Vaccines and Local Immunity, Siena, Italy.

10. Germanier, R., and E. Furer. 1975. Isolation and characteriza-tion of Salmonella typhi gal E mutant Ty2la: a candidate strainfor a live typhoid vaccine. J. Infect. Dis. 131:553-558.

11. Gilleland, H. E., M. G. Parker, J. W. Matthews, and R. D. Berg.1984. Use of purified outer membrane protein F (porin) prepa-ration of Pseudomonas aeruginosa as a protective vaccine inmice. Infect. Immun. 44:49-54.

12. Granoif, D. M., and R. S. Munson, Jr. 1986. Prospects forprevention of Haemophilus influenza type b disease by immu-nization. J. Infect. Dis. 153:448-461.

13. Hornick, R. B., S. E. Greisman, T. E. Woodward, H. L. Dupont,A. T. Dawkins, and M. J. Snyder. 1970. Typhoid fever, patho-genesis and immunological control, parts 1 and 2. N. Engl. J.Med. 283:686-691, 739-746.

14. Isibasi, A., E. Calva, V. Ortiz, M. Fernandez, A. Hernandez, andJ. Kumate. 1986. Vacunas contra la fiebre tifoldea a partir deantfgenos de membrana externa, p. 109-115. In Juan Garza(ed.), Avances en el uso de vacunas. Gerencia General de

J. CLIN. MICROBIOL.

ANTIBODIES AGA1NST PORINS IN TYPHOID FEVER 1645

Biolôgicos y Reactivos, Secretaria de Salud, Mexico City.15. Isibasi, A., V. Ortiz, M. Vargas, J. Paniagua, C. Gonzalez, J.

Moreno, and J. Kumate. 1988. Protection against Salinonellatvpihi infection in mice after immunization with outer membraneproteins isolated from Salmonella typhi 9.12.dVi. Infect. Im-mun. 56:2953-2959.

16. Joiner, K. A., E. J. Brown, and M. M. Frank. 1984. Comple-ment and bacteria: chemistry and biology in host defense.Annu. Rev. Immunol. 2:461-491.

17. Joiner, K. A., C. H. Hammer, E. J. Brown, R. J. Cole, andM. M. Frank. 1982. Studies on the mechanism of bacterialresistance to complement-mediated killing. 1. Terminal comple-ment components are deposited and released from Salmonellaminnesota S218 without bacterial death. J. Exp. Med. 155:797-808.

18. Joo, I. 1970. Present status and perspectives of vaccinationagainst typhoid fever. International Conference on the Applica-tion of Vaccines Against Viral, Rickettsial and Bacterial Diseaseof Man, Washington, D.C. Pan Am. Health Organ. Sci. Publ.226:339-341.

19. Kamio, Y., and H. Nikaido. 1977. Outer membrane of Sailmo-nel1cl typhitnuriiin identification of proteins exposed on cellsurface. Biochim. Biophys. Acta 464:589-601.

20. Karkhanis, Y. D., J. Y. Zeltner, J. J. Jackson, and D. J. Carlo.1978. A new and improved microassay to determine 2-keto-3deoxyoctonate in lipopolysaccharide of gram-negative bacteria.Anal. Biochem. 85:595-601.

21. Kishimoto, T. 1985. Factors affecting B-cell growth and differ-entiation. Annu. Rev. Immunol. 3:133-157.

22. Kuusi, N., M. Nurminen, H. Saxen, M. Valtonen, and P. H.Makela. 1979. Immunization with major outer membrane pro-teins in experimental salmonellosis of mice. Infect. Immun.25:857-862.

23. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)227:680-685.

24. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randail.

1951. Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193:265-275.

25. Lugtenberg, B., and L. Van Alphen. 1983. Molecular architec-ture and functioning of the outer membrane of Escherichfia coliand other Gram-negative bacteria. Biochim. Biophys. Acta737:51-115.

26. Montaraz, J. A., P. Novotny, and J. Ivanyi. 1985. Identificationof a 68-kilodalton protective protein antigen from Bordetellabronchiseptica. Infect. Immun. 47:744-751.

27. Osborn, M. J., and H. C. Wu. 1980. Proteins of the outermembrane of Gram-negative bacteria. Annu. Rev. Microbiol.34:369-422.

28. Rajagopalan, P., R. Kumar, and N. Malaviya. 1981. Immuno-logical studies in typhoid fever. 1. Immunoglobulins, C3, anti-bodies, rheumatoid factor and circulating immune complexes inpatients with typhoid fever. Clin. Exp. Immunol. 44:68-73.

29. Schnaitman, C. A. 1971. Effect of ethylenediaminetetraaceticacid, Triton X-100, and lysozyme on the morphology andchemical composition of isolated cell walls of Eschericlila coli.J. Bacteriol. 108:553-556.

30. Smith, K. A. 1988. Interleukin 2: inception, impact and impli-cations. Science 240:1169-1176.

31. Towbin, H. T., T. Staehelin, and J. Gordon. 1979. Electropho-retic transfer of proteins from polyacrylamide gels to nitrocel-lulose sheets: procedure and some applications. Proc. Natl.Acad. Sci. USA 76:4350-4354.

32. Wandan, M. H., C. Serie, Y. Cerisier, S. Sallam, and R.Germanier. 1982. Controlled field trial of live Salmonella tvphistrain Ty2la oral vaccine against typhoid: three years' results.J. Infect. Dis. 145:292-295.

33. Wang, L. I., and C. E. Frash. 1984. Development of a Neisseriameningitidis group B serotype 2b protein vaccine and evaluationin a mouse model. Infect. Immun. 46:408-414.

34. Westphal, O., and A. K. Jann. 1965. Bacterial lipopolysaccha-ride extractions with phenol-water and further applications ofthe procedure. Methods Carbohydr. Chem. 5:83-91.

VOL. 27, 1989